Skip to main content
Internet Archive's 25th Anniversary Logo

Full text of "Electrically scanning microwave radiometer for Nimbus E"

See other formats


^'i^ 



^^S^ 



r^ 



-I33SI3. 




Final Report 
for the 
ELECTRICALLY SCANNING MICROWAVE RADIOMETER 

FOR NIMBUS E 

(N&SA-CE-132812) ELECTRICALLY SCAHNIHG H73-32342 

HICEOSAVE E&DIOMEIEH FOE NIHBUS E Final 

Report (Aeroiet Eiectrosystems Co,)S^b%- p „ , 

nr Sii 50 CSCL UB Dnclas 

^^ *^'^^ G3/14 18209 / 

Prepared for 

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION 

Goddard Space Flight Center 

Greenbelt, Maryland 

1740FR-1 January 1973 



fffffo/er 

AIU»A, CAUFORNIA 



Final Report 
for the 
ELECTRICALLY SCANNING MICROWAVE RADIOMETER 

for 
NIMBUS E 

i740FR-l 
January 1973 



Prepared for 

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION 
Goddard Space Flight Center 
Greenbelt* Maryland 



In Response to 

Contract NAS 5-21115 

Article 4B 



Prepared by 

AEROJET ELECTROSYSTEMS COMPANY 
1100 West HoUyvale Street 
Azusa, California 91702 



FOREWORD 



This report covers work performed under Contract NAS 5-21 115 
during the period 19 February 1970 through 31 December 1972. The work 
was sponsored by the National Aeronautics and Space Administration's 
Goddard Space Flight Center, Greenbelt, Maryland. Program technical 
nnonitors for this contract have been Mssrs. Brice Miller and Dean Smith 
of NASA/GSFC. 



1740FR-1 Page ii 



ABSTRACT 



An Electronically Scanning Microwave Radiometer system has 
been designed, developed and tested by Aerojet ElectroSystems Company* s 
Microwave Systems Departnnent (formerly Aerojet General Corporation's 
Microwave Division) for measurement of meteorological, geomorphological 
and oceanographic parameters from NASA/GSFC*s Nimbus E satellite. The 
system is a completely integrated radiometer designed to measure the micro- 
wave brightness temperature of the Earth and its atmosphere at a microwave 
frequency of 19-35 GHz- Calibration and environnnental testing of the system 
have successfully demonstrated its ability to perform accurate measurements 
in a satellite environment* The successful launch and data acquisition of the 
Nin:ibus 5 (formerly Nimbus E) gives further demonstration to its achievement. 



1740FR-1 Pageiii 



CONTENTS 



SECTION 1 -INTRODUCTION - 

SECTION 2 -ELECTRICALLY SCANNING MICROWAVE 

RADIOMETER 

2, 1 General 

2.2 Applicable Documentation , . 

2. 3 Mass Model . 

2.4 Engineering Model 

2. 5 Protoflight Model 

2. 6 Flight Model 

SECTION 3 -BENCH TEST EQUIPMENT 

3. 1 General 

APPENDIX A - ENGINEERING TEST REPORT 

A. 1 Introduction 

A. 2 Equipment and Test Setup 

A. 3 Test Procedure 

A. 4 Results and Conclusions 



Page 
1-1 



2-1 

2-1 

2-1 

2-5 

2-7 

2-17 

2-21 

3-1 



A-1 
A-1 
A-2 
A-3 



1740FR-1 



Page iv 



ILLUSTRATIONS 



Figure No. Page 

2-1 Electrically Scanning Microwave Radiometer Mounted 

Aboard the Ninabus E Spacecraft 2-2 

2-2 Ninribus E Mass Model Number 1 .,-i 2-6 

2-3 Nimbus E Mass Model Number 2 . * 2-8 

2-4 RFI Susceptibility Corrections 2-10 

2-5 Data Record - July 26, 1971 2-11 

2-6 Analysis of Data Recorded - July 26, 1971 2-13 

2-7 ESMR Engineering Model - Component Side 2-15 

2-8 Electrically Scanning Microwave Radiometer 

Antenna Side 2-16 

2-9 ESMR Protoflight Model . . . . \ 2-22 

2-10 ESMR Flight Model . , 2-27 

3-1 BTE #1 3-3 

3-2 BTE #2 3-4 

3-3 BTE #2 - Rear View 3-5 

A-1 Thermal Balance Test, Eclipse Data, Top Heated . . A- 10 

A-2 Thermal Balance Test, Eclipse Data, Bottom Heated. A- 11 



1740FR-1 Page v 



Section 1 
INTRODUCTION 



The products developed under Contract NAS 5-21115 consisted of 
the Mass Model, Engineering Model, Protoflight Model and Flight Model of 
the Electrically Scanning Microwave Radiometer in addition to two each of a 
bench checkout unit capable of stimulating the input and interrogating the 
outputs of the radiometer and two bench checkout units capable of providing 
highly accurate targets for the radionneter. The following sections briefly 
describe each instrument and provides a reference for the applicable docu- 
mentation* 



1740FR-1 Page 1-1 



Section Z 
ELECTRICALLY SCANNING MICROWAVE RADIOMETER 



2.1 GENERAL 

The Electrically Scanning Microwave Radiometer basically consists 
of a scanning planar array antenna, Dicke microwave radiometer and post - 
detection electronics. The antenna is a planar electronically scanned array 
that scans in one plane. The antenna is coupled to the radiometer which com- 
pares the brightness temperature at the antenna to a temperature controlled 
hot (Dicke) load within the instrument. This comparison generates an analog 
voltage proportional to the temperature difference which is then quantized 
and further processed by digital circuitry included in the instrument. The 
output of the radiometer is a serial ten bit word for each beam position. The 
gain of the radiometer system is established by comparing a sky reference 
antenna (cold horn) to the hot load. The system automatically adjusts the 
gain to a norninal output. Calibration is accomplished by periodically com- 
paring the hot load with a known temperature load (ambient load) within the 
instrument* The instrunnent is then coupled to the Nimbus 5 spacecraft pro- 
viding a movable platform imparting motion in the non-scan plan of the antenna. 
This provides the necessary spatial displacement to generate two dimensional 
radiometric maps. Figure 2-1 depicts the Electronically Scanning Microwave 
Radiometer naounted aboard the Nimbus E Spacecraft. 

The specifications for the radiometer system are provided in 
Table 2-1, 

2.2 APPLICABLE DOCUMENTATION 

The following is a list of the documentation for the Electrically 
Scanning Microwave Radiometer. Not included in this list are monthly pro- 
gress reports, malfunction reports, screening reports and milestone reports. 



1740FR-1 Page 2-1 




ELECTRICALLY SCANNING 
MICROWAVE RADIOMETER 



Figure 2-1. Electrically Scanning Microwave Radiometer Mounted 

Aboard the Ninnbus E Spacecraft 



1740FR-1 



Page 2-2 



Table 2-1 
SPECIFICATIONS FOR RADIOMETER SYSTEM 

Receiver Specification 

Center Frequency 19. 35 GHz 

Bandwidth!. F, (nom, ) 10-150 MHz 

Bandwidth R. F. (nom* ) 300 MHz 

Mixer Noise Figure 5^ dB 

AT 1 5° K 

rms i. D r\ 

Absolute Accuracy 2.0°K 

Dynannic Range, Calibration 50-330° K 

Antenna 

Nadir 3 dB Beamwidth 

Scan Time 

Loss 

Polarization 

Sidelobe Contribution 

Scan Angle 

Total Experiment Weight 

Experiment Size 

Total Power Requirement 

The operation of the instrument is completely described in the Operation 
and Maintenance Manual, MW "PROC-8021. 

1740FR-1 Page 2-3 



<1.4° 




4. sec 




< 2. dB 




Linear 




< 7% 




± 50° 




67 lbs 




36" X 36" X 


6" 


41. Z watts 





Document 
No. 



Title 



Date 

Submitted 



1740 lM-1 



MW-PROC-8021 



SK1488-2001 
1488 Q-2 
1488 Q"l 

MW-SP-3001 D 
MW-SP-3003 N/C 
MW-SP~3020 N/C 
MW-PROC-8012A 



Still Documentary Photography 1-15-73 

Integration Manual for ESMR 7-7-72 
Nimbus E, October 1972 

Environmental Test Report ESMR 7-7-72 
Protoflight Model 

Environmental Test Report ESMR 7-7-72 
Flight Model 

Calibration Report ESMR Engineering 7-7-72 
Model 

Calibration Report ESMR Protoflight 7-7-72 
Model 

Calibration Report ESMR Flight Model 7-7-72 

Operation and Maintenance Manual 5-9-72 

Calibration Report, ESMR Antenna 5-4-72 
Acceptance Test, Flight Model 

ESMR Mechanical Interface Drawing 1-21-71 

ESMR List of Bulk and Raw Materials 7-20170 

Reliability Program Plan 5-12-70 

Inspection Flow Diagram 5-12-70 

Quality Program Plan 4-29-70 

Microwave Division Quality Control 4-20-70 
Manual 

Switch, Ferrite, RF 3-8-72 

Oscillator, Solid State, RF 5-7-70 

Integration Specification 12-1-71 

Acceptance Test Procedure, Ground 5-1-71 
Support Equipment Calibration/ 
Checkout Bench Test Unit No* 2 



1740FR-1 



Page 2-4 



Document 

No. 

MW-PROC-8013 
N/C 



MW-PROC-8014A 
MW-PROC-8015A 

MW-PROC-8016A 

MW-PROC-8017D 
MW-PROC-8018A 

MW-PROC-8019B 

MW-PROC~8022B 

MW-PROC-8033 
N/C 



Title 

Ground Support Equipment Operation 
Maintenance and Handling Procedure 
Calibration/ Checkout Bench Test 
Unit No. 2 

Shipping and Handling Procedure 

Acceptance Test Procedure ESMR 
Antenna 

Acceptance Test Procedure ESMR 
Receiver 

Acceptance Test Procedure ESMR 

Test Procedure Antenna Loss 
Measurement 

Acceptance Test Procedure Ground 
Support Equipment Electrical 
Checkout Bench Test Unit No. 1 

Environmental Test Procedure 

Test Procedure Liquid Nitrogen Target 



Date 
Submitted 

5-1-71 



12-15-71 
12-11-72 

6-15-71 

1-12-72 
1-17-72 

9-15-71 

11-5-71 
1-19-72 



2.3 



MASS MODEL 



The Mass Model constructed under Contract NAS 5-21115 was actu- 
ally the second Mass Model of ESMR. The first Mass Model was constructed 
under Contract NAS 5-11633. Structural deficiencies required the construction 
of the second Mass Model. The first Mass Model is shown in Figure 2-2. 

The second Mass Model was completed early in October 1970 and 
was vibration tested at Ogden Laboratories on October 13, 1970. The vibra- 
tion test revealed a resonance at 71 Hz, which was not acceptable. The Mass 
Model was fitted with a stiff ener channel which fit over the end of the model 
near the phase shifters. This aluminunn hat- shaped structure later supported 
the cable guides. 



1740FR-1 



Page 2-5 



-J 
o 



.^■^^'A^ii>^^ea^;fl'i?,^4^'i 





»I3 
1 



ANTENNA SIDE 



Figure 2-2. Nimbus E Mass Model Number 1 



RECEIVER SIDE 



The final vibration test of the model was performed at Ogden 
Laboratories on November 12, 1970. The results of this test were satis- 
factory and the Mass Model was shipped to GE/VFSC on November 19, 1970. 
The second Mass Model is shown in Figure Z-3. 

2.4 ENGINEERING MODEL 

Fabrication of the Engineering Model of ESMR was begun in April 
1970 and completed in February 1971, The environmental testing of the 
Engineering Model was accomplished during the month of March 1971* The 
environments included weight, center of gravity, RFI, thermal soak and 
vibration. The model was susceptable to RFI at 136* 5 MHz and was cor- 
rected through filtering of the mixer and intermediate frequency an^plifier. 
The unit also failed at +45 C* The failure of the radiometer to operate at 
+45 C was isolated and found to be caused by a temperature sensitive inte- 
grated circuit in the timing and control counter* After replacement of the 
I.e., thermal tests were repeated without failure. 

The Engineering Model was crated and shipped to GE/VFSC along 
with BTE #1 and BTE #2 on the 1st of May 1971* At GE, the hardware was 
uncrated and given a modified acceptance test by Aerojet-General, Micro- 
wave Division personnel. The ESMR measured radiometric temperatures 
to an absolute accuracy of 0. 5 K with a standard temperature deviation 

(AT ) of between 1.4 and Ue'^K. 

rms' 

The scan driver output voltages were naonitored through connectors 
J4 and J 5 on the ESMR. These voltages were recorded on a ten channel 
Brush recorder for future analysis by Microwave Division personnel. 

The ESMR was then integrated into the BIT (Bench Integration 
Test) facility for a compatibility test. For the results of this test, refer 
to the completed integrated test procedure (ITP 1420-NE-004) at GE/VFSC* 

During BIT testing, the following discrepancies were noted: 

• The grounds for chassis, signal and power were tied together 
in ESMR, The fix involved the separation and isolation of the 
grounds. 



1740FR-1 ^ Page 2-7 



-J 

o 




^miiiim: 




ANTENNA SIDE 



RECEIVER SIDE 



1 

00 



Figure 2-3. Nimbus E Mass Model Number 2 



• No digital B information on J 1-28 (output frame identification). 
The difficulty was traced to a miswire in module A- 7 and was 
corrected. 

• SAGC readout inoperative. Difficulty traced to broken wire in 
module A-7 which occurred during repair of the above, 

• Digital A output not synchronized to VIP (Versatile Information 
Processor) frame. Difficulty traced to an incorrect polarity of 
the MFP (Major Frame Pulse) input. This was corrected by 
reversing the MFP leads in the VIP* Later analysis has shown 
that the negative going MFP should have been presented at 
Jl-14 rather than J 1-31. 

• Digital A out of synchronization with VIP by one word (50 ms). 
The FID (Frame Identification) appeared on the first word rather 
than the last. This was corrected by a change in the TCC (Tim- 
ing and Control Counter). 

• Spurious operation of VIP synchronization and occasional word 
errors. This was traced to a malfunctioning flip-flop in Module 
A-7. The problem was corrected. 

• High current pulses on the -24.5 VDC input lines during turn-on. 

After the conclusion of BIT testing at GE/VFSC, the ESMR and as- 
sociated BTE were sent to NASA/GSFC for nnagnetic monnent testing. The 
test revealed residual nnagnetic moments which were probably caused by the 
magnetic field of the isolators in the RF section. The test concluded, however^ 
that these were inconsequential, and the system passed the test. 

The equipment was then returned to GE/VFSC. The ESMR was then 
crated and shipped back to AGC/MD with the two Bench Test Equipments re- 
maining at GE/VFSC. 

During the system inspection and test at AGC/MD it was found that 
all but one of the direct reading telemetry circuits were inoperative. Upon 
investigation, all the failures were traced to the transistors in the double 
Darlington circuits. 

A solution to the susceptibility of the ESMR to radiation at 136. 5 
MHz was incorporated in the radiometer. RFI filters were added to all leads 
going to the naixer-IF assembly as shown in Figure 2-4. 



1740FR-1 Page 2-9 



ADDED FILTEB^ y \ EXISTING 

/f \ / FILTER 



ru!SuT)°-"-SiKsm-r>~;^^5=^OfOA 



xt 



>-p^ I MIXER-IF P/N 371300_ j 



m 



-ADDED CAPACITOR 



Figure 2-4, RFI Susceptibility Corrections 

The new filters were installed in the bulkhead housing oi ttie radi- 
ometer section using the bulkhead as chassis ground. The 13.6 MHz applied 
via a tuned dipole at 1 meter from the ESMR antenna. The signal was pulse 
modulated at a 4 KHz rate with a 50% duty cycle at an average power level of 
1/Z watt. Change in the BIDEC count was minin^al. An average count (78 
samples) of 836. 4 and 836. 5 was recorded with and without RFI respectively. 
There was no increase in average deviation (AT) with the application of RFI, 

The ESMR Engineering Model tests were completed at AGC/MD in- 
cluding the measurement of current transients on the -24. 5 VDC input and the 
chart recordings of the phase shifter coil voltages for beam positions 5 through 
103 in the scan mode* The current transient for the 'RADIOMETER ON* po- 
sition was 4, anriperes. This was 2, amperes over the requirement of X- 
450-68-415. No discrepancies were noted for the beam position voltages. 

The Engineering Model was taken to Table Mountain for measure - 
nnent of the insertion loss of the antenna. 

The ESMR Engineering Model was sent to the General Electric 
Valley Forge Space Center on Monday, July 26, 1971. The unit was uncrated 
and inspected visually for any sign of shipping damage, both by G*E. and 
AGC/MD. No sign of damage was found and the unit was moved to the BIT 
(Bench Integration Test) area for acceptance testing. The unit was accept- 
ance-tested the following results as shown in Figure 2-5. 



1740FR-1 Page 2-10 



FUNCTION 


RUN 


Analog 



1 


2 


3 


4 


5 


3.91 


3.92 


3.91 


3.88 


3,91 


1 


1.32 


1.33 


1.28 


1.22 


1.23 


2 


1.51 


1.50 


1.51 


1.48 


1,52 


3 


1.54 


1.55 


1.55 


1.54 


1.56 


4 


1.54 


1.55 


1.55 


1.52 


1.55 


5 


I.' 42 


1.43 


1.42 


1.40 


1.42 


6 


1.55 


1.55 


1.56 


1.52 


1,56 


7 


1.10 


1.10 


1.12 


1.10 


1.15 


8 


4.08 


4.11 


4.13 


4.12 


4.15 


9 


4.37 


4,42 


4.44 


4.46 


4.48 


10 


4.53 


4.58 


4.60 


4.61 


4.65 


11 


4.60 


4.65 


4.67 


4.67 


4.70 


12 


4.33 


4.38 


4.40 


4.40 


4.44 


13 


4.25 


4.30 


4.32 


4.30 


4.35 


14 


4,47 


4.53 


4.54 


4.55 


4.58 


15 


1.21 


1.21 


1.20 


1.18 


1.21 


MUX 












1 


605 


602 


599 


596 


594 


2 


613 


609 


606 


601 


600 


3 


608 


606 


603 


603 


603 


4 


612 


607 


608 


603 


602 


5 


515 


515 


513 


516 


515 


6 


447 


447 


447 


447 


447 


7 


664 


664 


664 


664 


664 


8 


000 


000 


000 


000 


000 


ColdRef. Ave, 


947 


961 


959 


958 


922 


HotRef. Ave, 


150 


152 


150 


150 


149 


Variable Load 


99.7°K 


149. 9°K 


200. 0°K 


249. 7°K 


300, 0°K 


Variable Windov 


291. O^K 


292.0*^K 


293. 2°K 


295. 4°K 


296. 1°K 


Sky Load 


40. 09°K 


40. 07°K 


40. 07°K 


40. 03'^K 


40. 02°K 


Sky Window 


290. 9°K 


290, 9°K 


290. 9°K 


90. 7'^K 


290, 70K 


C (78 Sample) 


734 


599 


448 


299 


147 


ACrms 


5.21 


5.51 


5,79 


4.86 


5.78 



Figure 2-5. Data Record - July 26, 1971 



1740FR-1 



Page 2-11 



After acceptance testing, the ESMR was integrated into the BIT 
board for testing. The ESMR was tested functionally including the recording 
of current transients and scan driver voltages. The only discrepancies noted 
were a 4, 4 ampere current transient on the power line during radionneter 
turn- on and a voltage fluctuation in the analog telemeter outputs for linear 
array temperatures AA and CC. 

An 'RFI soak test^ was performed by using a ferrite rod antenna 
which was placed directly on the ESMR in six locations. The ESMR proved 
to be susceptible to RFI at 136* 5 MHz. It was agreed that the test was ex- 
tremely harsh and not really indicative of actual conditions. 

After the conclusion of BIT testing, the ESMR analog telemetry 
outputs were further studied to deternnine the cause of variation, A brush 
recording of the outputs revealed the change in voltage to be abrupt and 
between 5 and 10% change. Further analysis with a scope revealed a wave- 
form of approxinaately 0.25 volts, peak to peak, and approximately a square 
wave in shape with i*ounded corners* 

The data recorded at GE/VFSC on July 26, 1971 were further 
analyzed at AGC/MD with the results shown in Figure 2-6. 

The Engineering Model was once again returned to AGC/MD. It 
was inspected and found that no damage was incurred during shipping. A- 
nomalies on the analog temperature readouts experienced at GE were in- 
vestigated. Analog voltages were monitored on a strip chart recorder for 
five hours. No anomalies occurred. An analytical study was initiated to 
determine if the fluctuations experienced at GE could be caused by the ESMR. 

A test was conducted to simulate the temperature readout anomalies 
experienced at GE/VSFC* Each of the eight outputs of the temperature read- 
out circuits was passed through a ''break-out^' box and recorded on a strip 
chart recorder for an extended period of time. The break-out box was 
similar to that used at GE/VFSC when the anomalies in the readout circuits 
were witnessed. The strip chart records were thoroughly analyzed. An 
anomaly of the type experienced at GE/VFSC occurred with the circuit-as- 
sociated with the linear array "A'* temperature thermistor. In order to 

1740FR-1 Page 2-12 



I-" 
















Function 


Run 






1 


2 


3 


4 


5 




CalcTilated Cold ^ 
Reference Temp, ( K) 


103,51° 


103.49° 


103.49° 


103.49° 


103.48° 




Calculated Variable 
Load Temp. {°K) 


153,22° 


190.19° 


226.80° 


263.23° 


300. 02° 




Calculated Ambient 
Reference Temp. ( K) 


298,89° 


299.27° 


299.20° 


299.62° 


299.62° 




CalciiLated Hot Load 
(Dicke) Temp. ( K) 


334,3° 


334.29° 


334. 39° 


334. 27° 


334, 29° 




Gain (Counts /Deg. ) 


4.102 


4.164 


4.155 


4.149 


4.162 




AT (°K) 
rms ^ 


1.3° 


1.3° 


1.4° 


1.2° 


1.4° 




Ambient Load 
Abs, Accuracy ( K) 


-1.22° 


-1.56° 


-0. 87° 


-1.64° 


-0, 66° 




Antenna Port 

Abs, Accuracy ( K) 


+2.06° 


0.04° 


-0.28° 


-1.16° 


-1,06° 


tra 

(D 

t 


Figu 


ire 2-6. Analy 


sis of Data Recorded - July 26, 1971 





i^ 



return the ESMR to GE/VFSC in an expeditious manner, the two transistors 
(Q3 and Q4) of this readout circuit were replaced. The two old transistors 
were put into another readout circuit for additional testing. 

The ferrite switch thernnistor wiring was changed from analog 
channel '*0'^ to digital multiplex channel 3 and the waveguide thermistor 
wiring was changed from the multiplex channel 3 to analog channel ^'0". A 
change in the AGC CLEAR command circuitry was also made. When the 
AGC CLEAR conamand is executed, the SAGC will adjust the stepped attenu- 
ator to 3/8 of the maximum attenuation value rather than to the maximum 
attenuation value. This SAGC change causes the AGC CLEAR to read out 
399 on the digital output instead of 015 when commanded. 

The ESMR Engineering Model was sent to General Electric, Valley 
Forge Space Center on January 4, 1972. 

The transistors removed from the faulty double Darlington ther- 
mistor readout circuits were installed in a similar circuit in the laboratory* 
This circuit was tested for an extended period of time. The previously 
experienced anomalies could not be repeated. 

The Engineering Model was returned to AGC/MD from GE/VFSC 
with a report that there was no analog 6 or digital A output- The unit ar- 
rived February Zl and was examined, A broken wire was found in the analog 
6 circuitry which caused the no-output malfunction. A broken wire at the 
input and a broken wire at the output of the switch driver was also found. 
This caused the no-output condition for digital A, The ESMR was checked 
and found to be operating normally. It was returned to GE/VFSC on Febru- 
ary 22, 

No further incidences occurred on the model and it continued 
through the remaining tests at GE/VFSC and was then placed in storage. 

Photographs of the Engineering Model are shown in Figures 2-7 
and 2-8. 

A post vibration interface dimensional inspection yielded satis- 
factory measurements and the unit was transferred to TRW for thermal test- 
ing, 

1740FR-1 Page 2-14 



O 





I 



Figure 2-7, ESMR Engineering Model - Component Side 



I 

I 




Figure 2-8^ Electrically Scanning Microwave Radiometer 

Antenna Side 



1740FR-I 



Page 2-16 



2. 5 PROTOFLIGHT MODEL 

Fabrication of the Protoflight Model was begun in October 1970 and 
completed in November 1971. After completion a Systems Acceptance Test 
(MW-PROC-8012C) was performed. 

After successful completion of the Systems Acceptance Test (MW- 
PROC-8017C) and the RFI portion of the environmental test (MW-PROC-8022B) 
at AGC/MD, the protoflight unit was transferred to Ogden Labs for weight and 
center of gravity measurements and vibration test. During the vibration pre- 
test there was no data output, A faulty pin in a connector was suspected (how- 
ever, the problem was traced to a DC to DC converter problem described 
later). The connector was repaired, the pretest was successfully completed^ 
and the vibration test begun. 

During vibration in the thrust axis a wire on pin J 1-23 broke and the 
command buttons failed. This was corrected and the thrust axis was success- 
fully repeated. A low output during vibration in the transverse axis was traced 
to broken IC leads in the mixer/IF. 

During the transverse axis random test, a lack of radiometer out- 
put was traced to broken coils in the IF amplifier- This was corrected, the 
test was restarted, and a second failure occurred. This failure was caused 
by broken capacitor leads in the IF amplifier. The capacitors were a glass 
type, could not be epoxyed, and had been held in place with conformal coating 
material. It was deterimined that the coating compound used during a rework 
cycle had not been completely cured prior to the unit's being put back into test* 
Upon correcting the problem, the test was rerun and successfully completed, 

A post vibration interface dimensional inspection yielded satisfactory 
measurements and the unit was transferred to TRW for thermal testing. 

A functional evaluation was performed at TRW prior to the start of 
thernnal balance measurements, A broken wire on pin 16 of J 1 was discover- 
ed during this test and was corrected. The thermal balance measurements 
were successfully completed and the adequacy of the thermal design substanti- 
ated. The results of this thernraal balance test are included herein as Appendix 
A. 

1740FR-1 Page 2-17 



During the thermal vacuum test several computer errors during 
the first 12 -hour high temperature exposure forced cancellation and a re- 
start of the test- 

During the first low temperature exposure the multiplexed outputs 
were in an improper sequence. The system was returned to ambient* A 
SM54Lj72 flip-flop 1,C, in the A/D converter was determined to have failed 
when cold and was replaced* It was also noted during a subsequent ambient 
functional evaluation that the MUX parameters were occasionally out of sync* 
This problem was traced to a solder chip and a pinched wire on the Analog- 
to-Digital Converter *^A'' board* These two conditions occurred during the 
replacement of the SM54Xj72 I.C. 

The ESMR systenn was placed in a thermal chamber and thermally 
cycled to detect any additional thermal problems. It was noted that at high 
temperatures thfe Hot Load MUX count was high. This problem was traced 
to a broken wire on Pin 5 of P6t 

At cold tennperatures the lack of a radiometer output was traced to 
a faulure of the SHX-424 DC-DC converter to turn on reliably. An analysis 
of the circuit indicated that insufficient bias current was supplied to the 
switching transistors Q5 and Q6. In order for the converter to start reli- 
ably one of the two transistors, i. e. , Q5 or Q6, must supply enough current 
to saturate the transfornner* At low temperatures the reduced permeability 
of the transfornaer core naaterial and slower switching time of the output 
rectifiers required a larger annount of transistor collector current to cause 
core saturation. The base bias resistor was changed so as to provide suf- 
ficient base drive current to insure proper turn-on. 

Thermal vacuum tests were commenced on December 2, 1971* 
During the cold tenn^perature cycle it was noted that the multiplex paranaeters 
were intermittently out of sync. The test was continued while an investiga- 
tion of the cause of the sync problem was being made. During the hot tem- 
perature cycle of the test the nnultiplex parameters continued to lose sync 
intermittently* It was determined that this condition occurred due to false 
major frame identification pulses. The ESMR was returned to annbient tem- 
perature for further evaluation. The problem continued to occur intermittently, 

1740FR-1 ' Page 2-18 



This problem was traced to the thermocouple attached to a ferrite 
switch drive transistor and to the thermocouples, attached to the ESMR frame. 
These thermocouples were a part of the test equipment used by TRW to con- 
trol the thermal vacuum chamber tennperature during test- The thermo- 
couples were periodically sampled during the course of the test. When 
sampled, a large noise spike was generated through the thermocouple wiring 
which was coupled into the major frame identification pulse line fronn the 
Bench Test Unit. Noise spikes on the major frame identification pulse line 
caused the naultiplex readings to be out of sync. For the duration of the test 
the thermocouples were sampled only during times when ESMR test data were 
not being taken. The ESMR system thermistors were also utilized for cham- 
ber temperature monitor and control. 

The test was restarted on December 8* It was noted that the step- 
ped AGC count skipped* High ambient RFI was suspected to be the cause of 
this nnalf unction* A capacitor was added between the stepped AGC clock drive 
line and ground to eliminate this problem. 

The test was restarted on December 11 and continued without dif- 
ficulty until the thermal vacuum chamber lost vacuum during the 48 -hour high 
temperature test* The chamber was repaired and the thermal vacuum tests 
were completed on December 19 without further incident* 

Bench calibration and coil voltage tests were conducted at the MD/ 
AGC facilities and completed on December Z8. Preparations were made to 
move the ESMR to the Table Mountain facilities for antenna loss measure- 
ments (MW-PROC-8018) but had to be postponed due to bad weather and for 
roads to be cleared of snow* The ESMR was shipped to Table Mountain on 
December 31* This test was completed and the ESMR was returned to AGC/ 
MD on January 5, 1972* 

The ESMR was taken to the antenna range and digital pattern 
measurements were made. Upon examination of data during these nneasure- 
ments, it was noticed that there was an anomaly in the phase shifter coil 
current waveforms. The ESMR was returned to the Clean Room on Janu- 
ary 11* The problem was traced to a broken wire in the Increment Adder. 

1740FR-1 Page 2-19 



After rework, the coil current waveforms were proper. Upon analysis of the 
previous coil current waveforms it was noticed that this condition existed prior 
to initiation of environmental test. The effect of the broken wire was a 3% 
error in the positioning of the antenna beam. 

On January 13, the ESMR was returned to the antenna range for 
pattern measurements. Checkout of the measuring equipment with the ESMR 
continued. Measurements commenced on January 16. Patterns were mea- 
sured for the 78 beam positions, the Fail Safe position, and the no power 
condition. The patterns were acceptable* 

To insure that the loss measurements made prior to the Increment 
Adder rework were still valid, the ESMR was returned to the Table Mountain 
Facilities and new measurements were naade. Antenna losses were essen- 
tially unchanged. 

While at Table Mountain a problem developed in the hot lead. The 
trouble was traced to two broken wires in connector J6, These wires were 
determined to have been broken during the antenna pattern measurenaents of 
January 16 and were repaired. 

The ESMR was returned to AGC/MD on January 18 and the Bench 
Calibration Test (MW-PROC-8017) was successfully conducted on January 19- 
The covers were installed and Functional Test (MW-PROC-8033) was success- 
fully conducted on January 20. 

A final vibration test (MW-PROC-8034) was performed at Ogden 
Laboratories, This was followed by a repeat of the functional test at AGC/ 
MD. A GE compatibility test was conducted on January 2K All tests were 
accomplished successfully. 

The ESMR was packed and shipped to GE/VFSC on January 24, 1972, 
AGC/MD personnel accompanied the system to GE. Upon arrival at the GE/ 
GSFC facilities the unit was uncrated and inspected visually for any sign of 
shipping damage. No sign of damage was found. The functional test (MW- 
PROC-8033) was conducted by AGC/MD personnel aided by GE/VFSC per- 
sonnel* The ESMR functioned properly. 



1740FR-1 Page 2-20 



When the Protoflight Model was placed on the deployment mechanism 
at G,E. a slight interference occurred in the stowed position. The foana insula- 
tion on ESMR was slightly dimpled by screw heads on the deployment mechan- 
ism. It was also noted that the foam was too close (but not touching) to the 
mechanisna near the interface bracket area. The areas that were interfering 
were modified when the unit was returned to AGC/MD and the drawings were 
revised to incorporate the change. 

The Protoflight Model was integrated into the BIT (Bench Integration 
Test) facility at G^E. During the integration test, it was found that the model 
had a timing error in the digital output. A one word error occurred in every 
subframe (4 seconds) causing word 1 in the VIP (Versatile Information Processor) 
to read word 80 from ESMR, The Protoflight Model was then shipped back to 
AGC/MD at El Monte for failure analysis* 

The timing error was caused by the failure of the Microwave Divi- 
sion to retrofit the Timing and Control Counter to an existing B revision to the 
drawing. The revision was made to the unit eliminating the timing error. 

The Protoflight Model was then reassennbled and sent to Ogden 
Laboratories for a random vibration test to MW-PROC-8034, The model 
was then shipped back to GE/VFSC where it successfully passed the require- 
ments of integration testing on the BIT facility. After BIT testing, the Pro- 
toflight Model was placed in bonded stock at GE/VFSC* A photograph of the 
Protoflight Model is shown in Figure 2-9* 

2, 6 FLIGHT MODEL 

Fabrication of the Flight Model was begun in February 1971 and 
connpleted February 1> 1972. The ESMR was taken to the Table Mountain 
facility on February 3 and the Antenna Loss Measurement Test (MW-PROC- 
8018) was conducted. It was noted during the test that under warm ambient 
conditions the stepped automatic gain control (SAGG) setting, when cleared^ 
was below the normal inhibit value of 399* 



1740FR-1 Page 2-21 



o 



CD 
1 




Figure 2-9. ESMR Protoflight Model 



upon completion of the loss measurement test on February 4, the 
unit was returned to AGC/MD facilities and engineering evaluation tests were 
made concerning the SAGC- The gain of the video amplifier was reduced to 
allow a nominal setting of 479 when the SAGC was cleared. 

The Bench Calibration Test (MW-PROC-8017) was conducted prior 
to initiating the Environmental Test (MW-PROC-8022), The center of gravity 
was determined on February 6 and the unit was transported to TRW test fa- 
cilities the next day, and weighed on February 8. The ESMR weight was 
3K00 kg (68,35 lb). 

Vibration tests were initiated the sanne day. During the thrust axis 
and Y axis tests, step gain changes were noted* The ESMR was returned to 
AGC/MD for evaluation of this symptona. The problenn was traced to the Gunn 
oscillator (S/N 116). The faulty oscillator was removed and sent to the vendor 
for repair. Repairs consisted of optimization of the units internal impedance 
and replacement of the diode. The oscillator was returned to AGC/MD and 
installed on February 9- The ESMR was returned to TRW on February 10 for 
continuation of vibration test. While calibration of the vibration table was 
being performed with the naass model, excessive cross -axis vibration levels 
were experienced. It was decided that the Flight Model should go into the 
thermal vacuuin test while the table was being repaired and calibrated* A 
post vibration mechanical inspection per MW-PROC-8022 and the Bench Cali- 
bration Test was conducted on the Flight Model on February 10. Thermal 
vacuum testing (MW-PROC-8022) conramenced the next day. The tests were 
completed on February 18 and the ESMR was installed on the vibration fixture 
the same day. Remaining vibration tests were completed without incident. 
The ESMR was returned to AGC/MD where interface dimension and post vi- 
bration mechanical inspections were performed on February 21» 

The Bench Calibration Test was performed prior to shipping the 
ESMR to Table Mountain for post vibration antenna insertion loss tests. The 
ESMR was returned to AGC/MD on February 28. 



1740FR-1 Page 2-23 



The modification on the foam insulation as described for the Pro- 
totype was also made on the Flight Model. A check on the timing revealed 
the same timing slip as seen on the Protoflight and this too was corrected. 

The Flight Model was tested to the final Bench Calibration Test 
(MW-PROC-8017), Functional Test (MW-PROC-8033) and then a Final 
Vibration Test {MW-PROC-8034). During the Functional Test after vibra- 
tion, it was noted that analog #Z was not present. The cover was removed 
and the fault isolated to a broken wire (GSFC MRD02271). The wire was 
repaired and the Flight unit revibrated to MW-PROC-8034. After vibration 
another Functional Test (MW-PROC-8033) was performed. 

In addition to the above required testing, the Flight Model was 
tested for coil currents, current transients, continuity measurements, out- 
put voltages and susceptability at 136. 5 MHz (2 watts CW), All tests were 
satisfactory. 

The Flight Model was then shipped to GE/VFSC (March 27, 1972) 
after which it successfully passed both BAT (Bench Acceptance Test) and BIT. 

During the integration of the Flight Model to Nimbus E, field sup- 
port was requested at G*E. by the Ninnbus Project Office to determine the 
cause and possible cure for an RF interference problem with the Nimbus E 
Spacecraft, The ESMR was detecting a strong signal from the sensory ring 
while ESMR was in the stowed position. It caused saturation in several of 
the beann positions and erratic behavior in all others. The interference only 
occurred when any of the three S-band telemeter transmitters were turned on. 

Several tests were performed to determine if ESMR is susceptible 
to the beacon frequencies of 1702. 5, 1707. 5 and 2208. 5 MHz. These tests 
concluded that ESMR was not susceptible to these frequencies, but rather that 
the interfering source was within the bandpass of ESMR. Further testing re- 
vealed that the three S-band telenaeter transmitters were emitting energy fron^ 
four holes in each transmitter case which were drilled to prevent transmitter 
multipacking*. The four holes in one of the transmitters (S-band **A'') were 



^ A term used to describe an unwanted condition which causes frequency 
instability in tuned cavities. It results frona a combination of cavity 
size and atn:iospheric pressure. 

1740FR-1 Page 2-24 



covered with a copper tape to ascertain that the radiation was emanating from 
the holes. The test was conclusive in that the interference was no longer pre- 
sent when S-band *^A^* transnnitter was turned on and off* 

Large holes, 1/4" in diameter, were drilled by the manufacturer 
(Teledyne) in order to guarantee proper venting without need for vacuum analy- 
sis. The hole appeared to have a depth of about l/8'\ A 1/4'' hole appears to 
be beyond cutoff since according to Moreno's book on transmission lines, the 
formula for cutoff in the donninant TE. , mode is: 

Zrra 



'c 1.841 

where a is the radius of the circular guide. This calculates out to a frequency 
of 27.66 GHz for a 1/4" hole. Still, according to Moreno, the attenuation will 
be: 



a = 8.69 



J (i^) ^ - (^Y dB/unit length. 



Calculating this out for a frequency of 19 . 35 GHz reveals the hole to have an 
attenuation of 91-47 dB per inch or a total of 11.43 dB per hole. Since there 
are four holes, it can be assumed that whatever is on the inside gets attenu- 
ated by only 5. 43 dB before it gets to the outside world and ESMR. 

The radio frequency interference problem was resolved by shielding 
the holes in the spacecraft telennetry transmitters* The ESMR proved not 
susceptible after the fix was installed. 

The ESMR Flight Model continued through the spacecraft integration 
tests with no failure until thermal vacuum testing. During a retest phase of 
thernaal vacuum, the Dicke load thermistor readout circuit showed an inter- 
nnittent condition. The circuit involved places the physical temperature of the 
Dicke load in the digitial A stream to VIP on multiplex channel 5* The anomaly 
caused a shift of approximately 100 counts (out of 1023) on an intermittent basis. 
Through further investigation it was determined that the logical cause of the 



1740FR-1 Page 2-25 



disturbance is an operational amplifier (AR7, an M501B) in schennatic 1371284. 
Since the circuit involved is a multi- redundant circuit, the decision not to re- 
pair was nnade* 

On October 20, 1972 the ESMR radiometric output indicated a 
random increase in temperature. The random temperature changes lasted 
about 15 seconds starting approximately at 2239 spacecraft time* The data 
records showed the anomaly to be typical of an external RFI source, probably 
in the I*F. band of ESMR. The ESMR was in the anechoic chamber at GE with 
ESMR deployed and in full operation. 

A second and third occurrence of random temperature increases 
occurred at 0538 and 0610, spacecraft time, on October 23, 1972, The ap- 
pearance of the data was essentially the same as the occurrence on October 
20, but much shorter in duration. The spacecraft was outside the anechoic 
chamber with ESMR stowed for this test. Aerojet ElectroSystems was re- 
quested to send field service support in order to isolate the problem* 

While at GE/VFSC every effort was nnade by Aerojet and GE to 
isolate the problem. Several sources of external RFI possibilities were un- 
covered but no conclusions could be reached. The spacecraft was vibrated 
in the thrust (yaw) axis at full sinusoidal qualification inputs. A ring con- 
fidence test held after sine vibration on October 28 revealed no further anoma- 
lies. Field Service support was discontinued at this time with the assurance 
that GE would further investigate the anomaly after the spacecraft had finished 
vibration testing. 

The Flight Model continued through integration tests at GE/VFSC, 
The Nimbus E Spacecraft was shipped to the Western Test Range at Vanden- 
berg, California, on November 25, 1972, and launched December 1 1, 1972, 
at 2357 hours* A photograph of the Flight Model is shown in Figure 2-10. 



1740FR-1 Page 2-26 



-h1 

o 










itt 

.-^ 



-4 



Figure 2-10. ESMR Flight Model 



Section 3 
BENCH TEST EQUIPMENT 



3, 1 GENERAL 

The Bench Test Equipment for the ESMR consists of two basic units. 
One unit is the spacecraft telennetry simulation unit referred to herein as BTE 
#1 and the other is the variable cold load source, BTE #2. These two units 
provide the means to accurately calibrate the radiometer under conditions e- 
quivalent to actual spacecraft operation. Pertinent features of BTE #1 are as 
follows. 

The Telemetry Simulating Unit provides the ESMR, power, clock 
signals, digital A timing signals and command relay drive signals. The unit 
will also accept and display from ESMR, digital A (prinaary data), digital B 
and analog data* 

Power to ESMR is supplied by means of a regulated power supply 
with front panel nneters indicating supply voltage and radiometer supply cur- 
rent. Primary power is nominally Z4. 5 volts. Output voltage of the primary 
power supply may be varied (continuously) by means of a front panel potentio- 
meter from 0. to 40 volts. 

The clock signals supplied to ESMR are 2. 4 KHz square wave and a 
major frame pulse, 250 /isec wide occurring once every 16 seconds. Amplitude 
of these clock signals will be equivalent to the amplitude of the spacecraft clock 
signals (five volts nominally). 

Digital A timing signals "A^", '»Bj" and '^Cj" are supplied to ESMR. 
The "Bj'^ transfer pulse and ten **Cj, shift pulses occur every 2. 5 ms. The "A '^ 
enable pulse, enabling the ESMR Digital A readout, occurs once every 25 ms. 
Amplitude of these signals will also be equivalent to spacecraft Digital A ampli- 
tudes (five volts nonninally). 



1740FR-1 Pag^ 3_1 



Command relay drive signals are two simultaneous 24 volts ampli- 
tude pulses applied to the proper MA and MB lines. The pulses are 40 ms in 
duration initiated by front panel push-button switches, 

ESMR Digital A output is displayed by means of a BIDEC readout 
and a digital printer. A five -position rotary switch is used to select the 
Digital A data to be displayed and printed. Position No. 1 will display and 
print one of the 78 beam positions per scan (subframe). The particular beann 
position selected (per scan) is determined by a two decade thumbwheel switch 
operating in conjunction with position No. 1 of the rotary selector switch. Po- 
sition No. 2 of the switch will display and print each Hot Reference output. 
Position No. 3 will display and print each Cold Reference output. Position 4 
displays and prints continuously all ESMR Digital A outputs (40 outputs per 
second). Position No. 5 will display and print one of the eight multiplex para- 
meters of the ESMR frame (80th output of each subframe). The particular 
multiplex parameter displayed is selected by means of a one decade thumb- 
wheel switch operating in conjunction with Position No. 4 of the rotary se- 
lector switch. 

Digital B outputs are displayed by means of lamps within the push- 
button switches which activate the ESMR command relays. 

Analog telennetry outputs are selected by means of a 14-position 
rotary switch and displayed by a digital voltmeter. 

The Telemetry Simulating Bench Test Unit is housed in a standard 
19 -inch relay rack and is as shown in Figure 3-1. 

The variable cold load source provides variable temperature micro- 
wave terminations to the ESMR by means of two closed cycle cryogenic coolers 
which use heliuna as the coolant. One of these coolers is used to provide a 
simulated space temperature at the cold horn input into the radiometer while 
the other provides the means to calibrate the radiometer input. The loads are 
variable from approximately 40 K to 350^K. The load temperature is displayed 
on the face of the test equipnaent on the DORIC readout unit. 

The Bench Test Unit #Z is shown in Figure 3-2 and 3-3. 



1740FR-1 ^ Page 3-2 




Figure 3-1. BTE No. 1 



1740FR-1 



Page 3-3 




Figure 3-2. BTE No. Z 



1740FR-1 



Page 3-4 




Figure 3-3. BTE No. 2 - Rear View 



1740FR-1 



Page 3-5 



Appendix A 

ENGINEERING TEST REPORT 
EMSR-E ENGINEERING THERMAL BALANCE TEST 



A, 1 INTRODUCTION 

Prior to performing the ESMR-E thermal vacuum test (MD-PROC- 
8022B) on the protoflight model, an engineering thermal balance test was 
performed. This engineering test was conducted in the TRW 5* x 6' thermal 
vacuum chamber equipped with an LN- cooled shroud, A bank of heat lamps 
was used to heat first the cover side and then the antenna side of the ESMR. 
In each case the temperatures were allowed to reach steady state followed by 
an eclipse where the heat lamps were turned off. The ESMR was instrumen- 
ted with 41 thernnocouples to record the tenaperatures. The chamber pres- 
sure was maintained at 10 torr or lower throughout the test. 

A. 2 EQUIPMENT AND TEST SETUP 

The test was performed during Novennber 1971 in the TRW horizon- 
tal 5* X 6* thermal vacuum chamber located at TRW Systems, Redondo Beach, 
California. 

The chamber interior was completely covered by a black LN-^ cooled 
shroud capable of being maintained at -290 F, Chamber vacuum was main- 
tained by means of a diffusion pump and measured with an ion gauge. 

The ESMR was instrumented with 40 copper -constant in thermo- 
couples using 28 gauge wire. Eight other thernnocouples (T/C) were used to 
monitor the temperatures of the ESMR power cable bundles, the lamp fixture^ 
the collar and the LN^ shrouds. The T/C locations are listed in Table 1. All 
were attached with a small piece of tape or a small annount of RTV where the 
tape could not be used. ESMR power was supplied by the AGC/MD Bench Test 
Unit #1 Model 1371701-1. Hermetically sealed feed-throughs were provided for 
the T/C and power cables. All temperatures were recorded on two Esterline- 
Angus multipoint recorders. 

1740FR-1 Page A-1 



The heat lamp bank was mounted in the top of the horizontal chamber. 
The ESMR was positioned horiaontally with the irradiated surface 25 inches 
below the lamp frame. The ESMR was suspended from the top of the channber 
by thin aircraft cables connected to the four handling eyelets. For the inver- 
ted position with the array facing upward, small fiberglass extensions were 
fabricated and mounted to each of the four handling holes and had eyelets in 
them for the suspension cables. 

A rectangular collar or "picture frame** about 4 inches wide on a 
side was fabricated from sheet aluminum and covered with about 1/2*^ of super 
insulation* This collar was mounted around the upper edge of the ESMR to 
prevent irradiation from the heat lanrips from reaching the edges of the ESMR. 
This produced a more orbit -like situation with the external heat incident on 
only one surface and all other surfaces radiating to the cold shroud as they 
would to space. 

The power cables and T/C bundles were wrapped with super insula- 
tion to prevent them fronn having a signficant thernaal effect on the ESMR dur- 
ing the test* 

Five IR radiometers were used to measure the intensity and uni- 
formity of the irradiation from the lannps* Four radiometers were placed 
adjacent to the four edges of the ESMR and one was placed near the center. 
The output of the radiomieters was connected via a feed-through to voltmeters 
outside the chamber. Calibration curves were provided to convert voltage into 
heat flux. 

The lamp bank consisted of two separately controlled arrays. By 
monitoring the radiometer outputs and adjusting voltage to each lamp array 
some control over uniformity could be exercised. 

A. 3 TEST PROCEDURE 

The first phase of the test was with the ESMR mounted in the cham- 
ber with the covers facing upward and being irradiated by the Lamps. The 
chamber was evacuated to 10" torr or lower, LN2 was introduced into the 
shroud and the lamps turned on at a low intensity. As the temperature of the 



1740FR-1 Page A-2 



shroud dropped, the voltage to the lamps was increased* As the temperatures 
of the shroud and ESMR began to stabilize, the lamp voltage was adjusted to 
produce approximately a 100 F temperature on the covers while maintaining a 
fairly uniform intensity- Ternperatures were recorded on the multipoint re- 
corders constantly during the stabilization process and recorded on data sheets 
each half hour. Steady state was defined as less than 1 F change in one hour. 
Once steady state was reached the final temperatures were recorded and the 
heat lamps turned off to start the eclipse. The ESMR power remained on* The 
eclipse lasted one hour, during which temperatures were recorded* At the end 
of the eclipse the lamps were turned on to help return the ESMR to room tem- 
perature and the LN^ was purged from the shroud. After the ESMR and shroud 
returned to near ro6m temperature the channber was allowed to return to 
ambient pressure slowly. 

The ESMR was then withdrawn from the chamber and inverted so 
that the array faced upward toward the heat lamps. The above test procedure 
was then repeated* After 40 minutes of the eclipse the ESMR power was turn- 
ed off in order to approximate an orbital eclipse with power off. The duration 
of the power off eclipse was 30 minutes. 

The bottom heated test was concluded following the sequence de- 
scribed for the top heated test. 

This concluded the engineering balance test. 

A. 4 RESULTS AND CONCLUSIONS 

The steady state test temperatures are shown in Table I for both 
cases. The radiometer readings for the heat Lamps are shown in Table 11. 
Lamp intensity was adjusted to maintain approximately 100*^F on the electronics 
cover for the top heated case and 100*^F on the array for the bottom heated case. 
The 100 F temperature approximated the peak temperature for these locations 
based on orbital prediction. 

Maintaining these outer surface temperatures at the orbital peaks 
long enough for the entire ESMR to reach steady state with power on was con- 
servative from a verification standpoint since these surfaces reach these 



1740FR-1 Page A-3 



temperatures only periodically during orbit. Therefore, this test substanti- 
ated the high temperature performance of the thermal design of the ESMR 
since all components remained within acceptable temperature linciits* 

The ESMR-E computer model was used to simulate the ESMR in the 
chamber. Expected temperatures were calculated prior to the test and are 
also shown in Table I. The computer nnodel did not have sufficient nodes to 
provide a tenciperature for each T/C location* A uniform heat input was used 
in the calculations. Fairly good agreenaent between expected and measured 
data can be seen for the bottom heated case. For the top heated case the fact 
that some of the internal conn^ponents were calculated to be 40 to 45 F com- 
pared with measured temperatures of 90 to 95 F and 25 to 40 F calculated 
for the array face compared to a m.easure of 55 to 65 F indicated that some 
of the internal thermal paths of the. model and possibly the array surface pro- 
perties were incorrectly modeled. 

The temperature gradient across the phase shifters was monitored 
during the test. For the top heated case the gradient was 3 F fronn center to 
either end. This was well within acceptable limits. For the bottom heated 
case the maximum gradient was 17 F between the center and the -Y end of the 
phase shifters (87 center, 70 end) and 8 F between the center and the -l-Y end 
(87 center, 79 end). (All axis designations refer to spacecraft axis system,) 
This rather high asymnnetrical gradient can be explained by the fact that the 
lamp intensity was 7% lower on the -Y side compared with the +Y side and that 
the DC/DC converter which was 133 F for this case was located on the +Y side* 
These two factors combined to cause asyminnetric heating. The temperature of 
133 F for the DC/ DC converter had a stronger influence on this gradient during 
the test than it will in orbit since the nnaximum orbital temperature is expected 
to be 117^F. 

Figures I and II show the temperature histories for various sections 
of the ESMR during the two test eclipses. Figure I shows the 60 minute eclipse 
for the top heated case. Figure II shows the 40 minute eclipse for the bottom 
heated case followed by the 30 minute eclipse with ESMR power turned off. The 
power -on portion of this eclipse is shown with solid lines and the power -off 
portion is shown with dashed lines. 



1740FR-1 Page A-4 



The deployed position in orbit is most clearly simulated by the 
bottom heated eclipse. In the deployed position in orbit the bottom or array 
face receives solar ^radiation just prior to passing into the earth* s shadow* 
This produces array temperatures near 100 F as in the bottom heated test 
case* The orbit occulation time is approximately 35 minutes. Figure II 
shows that the ESMR electronics section decreased in temperatures approxi- 
mately 28 F during the first 35 minutes of the eclipse in the chamber. The 
phase shifters decreased about 24 F for the same period. This data can be 
compared with the predicted decrease for orbit occulation of about 15*^F for 
each section. However, in orbit the ESMR absorbs approximately 75 watts 
due to earthshine during occulation which tends to decrease the temperature 
decline. On this basis the temperature decrease of 24*^ to 28*^F in the cham- 
ber indicates that the thermal design will perform satisfactorily during orbit 
occulation with power on. 

A power -off eclipse test starting with steady state temperatures 
was not performed. However, after the 40 minute power ^on eclipse for the 
bottom heated case the power was turned off and a 30 minute power -off 
eclipse was performed. The start temperatures for this eclipse was within 
10 F of those calculated for orbit with the exception of the DC/ DC converter. 
The DC/ DC converter did not have sufficient time to cool before starting the 
eclipse. The electronics, radiometer and phase shifter sections registered 
a temperature decrease of about 30°F. Orbital calculations showed a decrease 
of about 14 F for these areas. This is similar to the power -on eclipse in that 
the presence of the earthshine would account for the difference between the 
orbit and chamber eclipses. Also, during the chamber eclipse the temper- 
atures did not fall below acceptable limits. Hence, the thermal performance 
of the ESMR during a power -off eclipse is considered to be satisfactory. 

Comparison of the expected test calculated temperatures with the 
measured data shown in Table I indicates that upgrading of the computer model 
was desirable before final orbital temperature predictions were made. This 
was especially evident in the top heated case in the area of the electronics and 
radiometer sections. By examining the thermal paths in these areas it was 
found that conduction paths formerly considered to be insignificant must be 



1740FR-1 Page A-5 



included in the model. Also, other conduction paths and some radiation paths 
had been estimated too low. Finally, it was determined that the effective e 
of the array face was actually lower than that used in the calculations. It 
appears now that thd € of the array is about 0. 50 instead of the 0. 57 that was 
based on calculations and some inconclusive laboratory nneasurements. 

With a computer model incorporating the above changes, final 
temperatures were calculated for the test conditions. These results are 
also shown in Table I, It can be seen that all areas are within 10 F of the 
test temperatures. 

As a result of this thermal balance test the corrected computer 
model is considered to be sufficiently accurate for predicting final orbital 
flight temperatures for ESMR-E* 

Final orbital temperature calculations have been nnade using this 
refined model and are presented in the appendix. For the deployed position 
all temperatures are well within operating limits for all of the equipment. 
The low temperatures for the deployed power -off condition are well above 
the minimum start-up requirements* All temperatures for the stowed power - 
off condition are safely within the range that could have an effect on any of the 
ESMR components. The thernaal AT*s both across and over the honeycomb 
panel are not expected to cause warpage that could significantly affect the 
ESMR. 

Also shown in the appendix are the final surface properties used 
in the orbital calculations. These data were determined from both thermo- 
physics measurements and the thermal balance temperature data. 

Based on both the test data and analytical results it is evident that 
the ESMR-E will experience acceptable temperature limits throughout the 
Nimbus -E naission. 



1740FR-1 Page A-6 



-0 

■ o 



T/CNo. 

Z 

3 



S 

9 
10 



14 
15 
16 

17 

1 9 

" zh 

. 21 

• 22 

z^ 

T * 

25 
2d 
27 
23 
29 
30 
31 
32 

> 

r 
-J 



Tabic I 
Thermocouple Locations and Steady State Temperat-arc Data 



Location 

Outside phase shifter cover, center^ 

Outside phase shifter cover, r Y a>ds 

CutsLuc phase shifter cover, -Y axis 

O^tsi^lc elect ror-ics cover, center 

Cutsidc electronics cover, ever DC -DC converter 

Outside electronics cover, over Gunn oscillator 

inside of side rail next to pr^se shifter, {-i-Y} 

End ph:isc shiiccr coil, *5- * 

Center ohasc shifter coil, (^045) 



nhasc sr 



tcr coil. 



Inside c: side rail nc:ct to prjise shifter {-Y} 

L-iside ircr.t cress rail center, center 

Circuit b'jard A 15 (center board) 

Sv.'itch driver transistor 

DC -DC Convert or, top 

Circuit beard A6 

Pov--cr trans is t=*r mounted oh center open rail 

Cur-n Cscille^tor 

Z-r.T\ Cscilliitor Radicator 

y^Qw ^Z-'DC Board between capacitors 

Center cross rail between thermaUoys 

Mixer, top ' < , 

Cold horn 

P.car deck, + Y 

Cellar, front 

Outside rear cross rail, center 

Rear cec>:, locating block 

Rear edge, center 

Outside of team, next to phase shifters, -Y 

O-utsidc c: mour.ting, interface, +Y 

Outbids of foam, side, rca.r, -^Y 

Ar^t^^^ :=,ce, -X, tY corner (base) 



Ton Heated 



S.S. T/C 

Rcadintis 

102 
101 
96 - 
93 
ICl 
94 
77 
S4 
87 
-. S4 
79 
S3 
92 
110" 
115 
94 
84 
90 
91 
, 120 
91 
94 
68 
127- 
-50 
S6 
'75 
50 
65 
78 
76 
* 54 



Cz 



:t ed 



ilatcd 
Tom OS 



Calc\Alatcd 
Final Tcm^c: 



110 - 115 
I 

90 - 95 



90 - 95 

95-100 
90-95 

40-45 
40 - 45 

105 - no 

40 - 45 
40 - 45 



40 - 45 
40 - 45 

115 - 120 



25 - 30 



* s Defective thermistor connection 



104 

t 

107 



! 

84 

89 
84 

83 

33 

122- 
S3 
S3 



83 
130 



53 



Bottom Heated — * 



S.S. T/C Ca: 
Rcadin::s Z:--z 



64 

67 

68 

76 

S5 

75 

85 

79 
. S7. 

70 

65 

93 
100 
113 
133 . 

99 

99 
109 
109 

100 

112 

S3 

35 

-100 

Z5 

■ S4 

63 

70 

36 

91 

104 



45 - 50 



60 - 6= 



70 - 75 



95 - 100 



140'- 143 
95 - 100 
95 - 100 



95 - 100 

95 - 100 

10 - 15 



95 - 104 



i 

£4 
90 
SO 

99 



99 



99 
28 



101 



o 



T/CNo. 

33 

34 
35 
3d 
37 
33 
39 
■ 40 



43 

46 
43 



Table I {Cont. ) 

Thermocouple Locations and Steady State Temperature Data (cent. } 



■Location 

Antenna face. -X, Center (Base) 

Antenna face, center* center (base) 

Antenna face, center, center (surface) 

Anier.na face, under DC-DC (base) 

Antenna face, under rear deck (surface) 

Ir.siclc side rail, next to DC-DC . 

Under DC-DC Mounting Screw 

Tront cc!^e cap^ center 

Cc-:cr r-^asc shi'tcr coU («009) 

power Cziblc, tY 

Power cable > -V 

Collar, ris-^ s'-^- ^* chamber 

Eack \;-all of shroud 

Main shrcud, too £ro::it 

D^or shroud 

JLanr.o iL>:turc 



Tod Kcatcd ^ F 



Botton-i Heated 



0-, 



s.s. 


T/C 




Cal 


culated 


Calculated 


S,S. T/C 


Calculated C 


alrulatcd 


Read: 


.n-s 


^ 


Koccted Tennps ' 
25-30 , 


Final.Temos 
58 


'Rea diners • 
ICO 


Z>r:;ected Temos Pi: 


- J.1 Tcn'.tJS 


63 


95 - 104 


i * 


63 






35 - 


40 


53 


100 




1C5 


57 






30 - 


35 ' 


58 


103 




1 '>^3 


63 






35 - 


40 


53 


110 




1C5 


17 






15 - 


20 


34 ' 


94 




^99 


95 








• 




104 


T 




S8 












113 


' 




65 








* 




* 81' 






86 






95 - 


100 


S9 


. 90 


70-75 


S9' • 


75 












53 ■ 






* 73 








. 




■ 54 






-100 












-137 






-290 












-300 






-295 










■ 


-295 • 






-273 






' 


^ 


' 


-295. 




- , * 


• -4 












-6 ■. 







ft) 



00 



Table II 



-J 

o 
*^ 
SO 



STEADY STATE IR R/^DIOMETER READINGS 





Top Heated 










Radiometer No. 


1 


2 


3 


4 


5 




Serial No- 


15 


16 


17 


21 


Special 




Location 


Front 


Right (-Y) 


Rear 


Left {+Y) 


Center 


(Axis location refers 
to S/C axis system) 


Voltage 


1.250 


1.171 


1.231 


1.179 


1.258 


' 


Absorbed energy 
BTU/Hr ft^ 


320 


295 


306 


292 


354 




Incident energy 
BTU/Hr ft^ 


352 


324 


336 


321 


390 


Ave. = 345 




Bottom Heated 










Location 


Front 


Right (+Y) 


Rear 


Left (+Y) 


Center 




Voltage 


1.085 


1.022 


1.068 


1.027 


1.084 




Absorbed energy 
BTU/Hr ft' 


221 


212 


217 


193 


216 




Incident energy 
3TU/Hr ft" 


242 


233 


239 


218 


238 


Ave. = 234 



p 



^ 



THERMAL HAIANCE TI'ST 
ECI.1[>3L DATA 
TOr MlATf-.D 



LU 

Hi 

I— 
< 

LU 
Q. 

? 

UJ 







-20 



-^0 l- 



DC-DC 
CONVERTER 



POWER ON 




ARRAY 
FACE UNDER 
EQUIPMENT 



ARRAY FACII 
UNDER REAR DECK 



20 --^ ^ 
TIME-MINUTES 




(•[CURE A- J 



1740FR-1 



]^if',«' A- 10 



. TlirUMAI ISAIAHCJ- Tl:ST 
1^ 1 irsi DATA 
h()l|(.)/.A til All 1) 



120 



---,,./■■• 



/ DC-nc co;-iVi:kitR 



100 



00 



60 



[-:i.l:"CT!U.:)MlCS AND 
RADlOMLTtR SrCTION 



LU 

CrJ. 



LU 

2 



40 



I- 



20 




AI'.iiAY !ACH 

UNi.ji:i; [-.QUIP 




N. 



rtiASI! SHIFTKR 
COVER 



. — ARRAY [-ACE 
/ UNDER RHAR DHCK 



•^^ 






1 



.„L 



J 



-20 



-/JOL 



10 



20 30 40 

TIME -MINUTES 



50 



60 



70 



FIGURE A -2 



1740FR-1 



P.,j.'.f A- 11 



Orbit. - Max, Mln. 



AT Acror;s J''ha,^t^ rAiiftcr 

Pha.sc Rlriflcr Cover 

Electronics: ami Radiometer- Section 

Electronic ^5 aiid Ro^dioincter Cover 

DC -DC Converter 

Antenna Face 

Max A'J' ocross Honeycomb 

Max aT over Antenna Face 



Diiployod 


Dcp](iyi:(! 
Power OiC 


Slowed 
Fowrr Oi",r 


86 - 65 "f 
(3C)-1b"c) 


63 - 'II^F 
(17-5''C) 


61 - ^13"f 
(16-6"C} 


7'^F 
(4"C) 


5"f 
(3''C) 


5^'F 
(3"C} 


111 - 50^'f 
(44 - 10"g) 


90 - 28 "f 
(32 ~ ~2'-*C) 


81 - 28"f 
(27 - -2"C) 


84 - 65 "f 
(29 - 18"C) 


57 - 39*^F 
(14"4"C) 


46 - 35 "f 
(8 - 2"C) 


103 - 49*^}? 
(39 - 9 C) 


82 - 25 °F 
(28 ~ '-4"C) 


69 - 23"f 
(21 - -5°C) 


117 ~ lOsV 
(47 - 41 "C) 


53 - 41 V 
(12 - 5°C) 


44 - 37°F 
(7 ~ 3^C) 


85 - 41°F 
(29 - 5°C) 


66 - 30°F 
(19 - -iV.) 


56 - 30°F 
(13 - -i"G) 


9°F 
(5^0 


10°F 
(6°C) 


6V 
(3"C) 


16^F 
(9°C) 


(6°C) 


16°F 
(9"C} 



FINAL SURFACE FIN.ISH PROPERTIES 



Covers: D4D Ahiminum Paint 



o - 0,245 
s 



c = 0.280 
Ajitcima Face: 



0- = 0.66 
s 



e = 0.50 

Side Edges: Foam Insulation 
1740FR-1 



Page A- 12